OpenAI Expands Free Frontier Model Access to as Many as 200,000 Scientists
We’re expanding free access to our frontier models for up to 200,000 scientists worldwide, with training and support. Applications open Nov. 23. By default, we don’t use data shared through the program to train our models.
Context
A much larger free access program could put frontier models in the hands of working scientists at no cost, which may speed discovery while also spreading one lab's tools through academic work.
OpenAI posted on November 16 that it is expanding free access to its frontier models for up to 200,000 scientists worldwide, with training and support, and that applications open November 23. The company said that by default it does not use data shared through the program to train its models. In a second post the same day, OpenAI said a study of more than 1,800 researchers in the existing program found they reported nearly doubling their productivity with AI.
Deep dive
OpenAI said it is expanding the ChatGPT for Academic Researchers program to offer free access to its frontier models for up to 200,000 scientists, with training and support. Applications open November 23. In a blog post the same day, the company said the program began as a July pilot and that the next round will award 50,000 new grants on a first come, first served basis to eligible applicants, replacing the earlier lottery. Each new grant covers the recipient and up to two collaborators and provides ChatGPT Plus, which OpenAI said can support up to 150,000 researchers in this round. The 200,000 figure is the goal for the program as a whole, double the target announced when the pilot launched.
The blog said 20,000 researchers in 33 countries have joined since July. OpenAI also said that by default it does not use data shared through the program to train its models. The pilot had offered ChatGPT Pro, but the company said most participants would have stayed within Plus usage limits, so new grants use Plus in order to reach more teams.
A companion post and paper, Science at the Frontier, covers a study of more than 1,800 participants. OpenAI said scientists reported nearly doubling their productivity with AI, and that surveyed participants estimated the same work at the same quality would have taken nearly twice as long without AI. Those results come from OpenAI's own economic research team and from participant reports, not from an independent audit.
What to expect next
The next test is whether the November 23 application window fills the 50,000 new grants, and how OpenAI chooses among eligible applicants once that cap is hit. Watch for later public details on which fields get access, how Plus limits work in practice for heavy research use, and whether the productivity claims hold up outside OpenAI's own study. The company said it still aims to reach 200,000 researchers across the life of the program.
